| VaR 101 |
|
|
| 04/03/2007 | |
|
Ahmed Talhaoui, CFA and Head of Portfolio Solutions at Barclays Global Investors, answers this week's 101 question on VaR. 1. What is Value at Risk (VaR)? VaR stands for Value At Risk. It was first introduced by a specific team within the risk group of JP Morgan which spun-off into a company on its own called Riskmetrics. It is a very popular technique for measuring market risk that has been widely used by various financial institutions, including central banks. It relies on simple mathematical assumptions around the returns of markets and uses statistics to assess the probability of a portfolio to outperform or underperform by a certain level over a certain horizon. The methodology is relatively simple, it assumes the returns of the portfolio are normally distributed and the VaR is calculated with a covariance matrix derived from observed historical data. Because the methodology is accessible, most of the investment banks and asset managers decided to use this methodology. A majority of central banks for instance will actively monitor the VaR of the mandates they have allocated to external managers. Some institutions even use VaR as a metric for some regulatory requirements. Another development of VaR is that the methodology has been extended through simple adjustments to include instruments that were not supported in the past, such as derivatives. 2. Can VaR be a measure of volatility in an adverse standard deviation event? VaR is not a measure of volatility in an adverse standard deviation. VaR is simply a conservative multiple of a standard deviation. As such, some institutions have questioned the validity of VaR in "extreme" events where the magnitude of deviations happens to be much higher than what the mathematical modelling would imply. However, no simple methodology has been widely accepted as the replacement of VaR so many pension funds keep on measuring it, while using in addition other risk management tools such as scenario analysis and simulations. 3. What are the advantages and risks of using VaR as part of a fund's investment strategy? The first advantage of VaR is its simplicity. It is relatively straightforward to build a decent VaR analysis framework and most of the analytics providers offer VaR solutions. Also, because of specific properties of the covariance analysis, VaR analysis can offer a contribution of risk per strategy, which is very useful for a fund running multiple strategies. Some products could be running very low VaR and be mediocre. VaR has to be measured against specific risk limits and targets which are tied closely to the expected outperformance and the flexibility of investing. There are many shortfalls in VaR but if the historical data used is not flawed, it can flag interesting diversification patterns within a fund. Lastly, monitoring VaR as part of a fund's investment strategy is a very neutral, non-emotional disciplined risk control that can be implemented outside of the investment team. |
|
Articles of the same Serie : 101 |
|
© Copyright 2008 bfinance. This document is for your personal non-commercial use. Any further copying, reproduction, distribution is strictly prohibited. To obtain permission please contact This e-mail address is being protected from spam bots, you need JavaScript enabled to view it